study objective: To evaluate the efficacy and side effects of niacin therapy in dyslipidemic individuals. design: A retrospective analysis of patients' charts. setting: An outpatient referral-based clinic specializing in the treatment of lipid disorders. patients: All patients with dyslipidemia treated by niacin (n = 82) at the Atherosclerosis Detection and Prevention Clinic during 1987 to 1990, including a subgroup of 17 dyslipidemic heart transplant recipients. results: Niacin was well tolerated in 83% of the nontransplant group (n = 65) at an average dose of 2.5 ± 0.9 g/day. Similar beneficial lipoprotein effects were found in the transplant and nontransplant patients. The high-density lipoprotein cholesterol (HDL-C) response to niacin therapy was independent of the baseline HDL-C level. In the transplant group, 11 patients (65%) discontinued treatment, primarily because of hyperglycemia; this was especially prominent in those patients with pretreatment diabetes mellitus. Of the 15 patients using sustained-release niacin, eight cases of hepatitis were recorded, some during therapy with relatively low niacin doses. Several different sustained-release preparations were responsible for this phenomenon, suggesting that the cause was not a contaminant in the preparation. No cases of hepatitis were documented in the 67 patients using regular niacin. One case of hepatitis was recently observed in a patient who switched from one type of regular niacin to another, however, we have data to suggest that the substituted preparation was not an immediate-release niacin. A familial predisposition to hepatitis is suggested by the occurrence of this side effect in identical twin brothers and two sisters. A pharmacy survey disclosed that most pharmacists are unaware of the relationship of sustained-release niacin to hepatitis, have a negative impression of regular niacin, and do not stock this formulation. Finally, we found that in this small sample of patients, niacin used with lovastatin is a particularly effective drug combination and appears to have few side effects beyond those seen with niacin alone. conclusions: Our experience supports the fact that regular niacin is a useful lipid-modifying drug. When used appropriately, patients can usually tolerate adequate doses for prolonged periods and achieve meaningful results. However, this requires a certain amount of physician skill and patient motivation. The use of sustained-release preparations to overcome this problem can lead to harmful consequences and should only be done under strict medical supervision. In our opinion, the availability of sustained-release niacin as a nonprescription drug is unjustified and should be reexamined. Finally, we have observed that reduction of very-lowdensity lipoprotein cholesterol (VLDL-C) with niacin alone leads to an elevation in low-density lipoprotein cholesterol in many patients; this indicates to us that the mechanism whereby niacin lowers VLDL-C and total cholesterol is not solely the result of a decreased synthesis of VLDL-C. © 1991.